In the April 2010 issue of this journal, Date et al. expressed concern over the slow scale-up in low-income settings oftwo therapies for the prevention of opportunistic infections in people living with the human immunodeficiency virus: co-trimoxazoleprophylaxis and isoniazid preventive therapy. This short paper discusses the important ways in which policy analysis can be of use inunderstanding and explaining how and why certain evidence makes its way into policy and practice and what local factors influencethis process. Key lessons about policy development are drawn from the research evidence on co-trimoxazole prophylaxis, as suchlessons may prove helpful to those who seek to influence the development of national policy on isoniazid preventive therapy and othertreatments. Researchers are encouraged to disseminate their findings in a manner that is clear, but they must also pay attention tohow structural, institutional and political factors shape policy development and implementation. Doing so will help them to understandand address the concerns raised by Date et al. and other experts. Mainstreaming policy analysis approaches that explain how localfactors shape the uptake of research evidence can provide an additional tool for researchers who feel frustrated because their researchfindings have not made their way into policy and practice